Paper #1:
Putting Things in Context: Challenge on Context-Aware Movie Recommendation

The document discusses the context-aware movie recommendation challenge (CAMRa 2010). A total of 40 team submitted papers out of which 10 were chosen to be presented at the event. The rules of the challege provided the following: two data sets (train set and test set) to be evaluated under 3 statistical rules in order to determine how good each recommendation was. There were 3 recommendation tracks to choose from: a- time of the year and special events, b- social relations of users, and c- user's (implicit) mood.

The closing section of the paper briefly describes some of the results obtained by the ten chose papers. Out of those, there are two particular papers my team would be interested on; they are related to the social relation of users track. The social approach on [10] was based on matrix factorization. A feasable approach if we choose to do the recommendation calculations on server side. Finally, [11] presented a kNN approach based on linear combinations of similarity between users.